Differentiation of food pathogens using ftir and artificial neural networks

M. J. Gupta, Joseph Maria Kumar Irudayaraj, C. Debroy, Z. Schmilovitch, A. Mizrach

Research output: Contribution to journalArticlepeer-review

Abstract

FTIR absorbance spectra in conjunction with artificial neural networks (ANNs) were used to differentiate selected microorganisms at the generic and serogroup levels. The ANN consisted of three layers with 595 input nodes, 50 nodes at the hidden layer, and 5 output nodes (one for each microorganism or strain). Ten replications of each experiment were conducted, and 70% of the data was used for training and 30% for validation of the network. Results indicated that differentiation could be achieved at an accuracy of 80% to 100% at the generic level and 90% to 100% at the serogroup level at 103 CFU/mL concentration.

Original languageEnglish (US)
Pages (from-to)1889-1892
Number of pages4
JournalTransactions of the American Society of Agricultural Engineers
Volume48
Issue number5
StatePublished - Sep 2005
Externally publishedYes

Keywords

  • ANN
  • Differentiation
  • Food pathogens
  • FTIR spectroscopy

ASJC Scopus subject areas

  • Agricultural and Biological Sciences (miscellaneous)

Fingerprint Dive into the research topics of 'Differentiation of food pathogens using ftir and artificial neural networks'. Together they form a unique fingerprint.

Cite this